Offshore wind turbines are subject to cycles of loading and unloading from their environment i.e. wind and waves. This cyclic loading can significantly weaken the foundations over time leading to a reduced loading capacity when compared with a static load of the same magnitude. Structures can fail due to the build-up of plastic strain in the soil over time through successive loading cycles. In addition, there are failure mechanisms caused by liquefaction (where pore water pressure builds up and soil friction de-creases). Through laboratory experiments this project will seek to develop a constitutive model of sandy soil under cyclic loading. This will provide a better understanding of the behaviour of sandy soils with cyclic loads applied. Therefore, it should be possible to improve the design of offshore wind turbine foundations. This is of particular significance to the UK as much of the seabed surrounding Britain has at least partial presence of sandy layers, with some areas comprised almost entirely of sandy soil e.g. southern North Sea. Improvements in the design of offshore wind turbine foundations should lead to material and cost savings, helping to drive down the levelised cost of energy.
The project methodology will be focussed around gathering results from cyclic triaxial tests. Triaxial tests are a standard industry test used to measure the mechanical response of a soil sample to prescribed stress and/or strain boundary conditions. The cyclic triaxial tests will apply computer controlled cyclic loads to a range of samples prepared with different initial densities and stress histories. The samples will be prepared in the lab, as opposed to testing cored samples from the field due to the tendency of sandy soils to break apart when loads are released. The results will be used to develop and calibrate a constitutive soil model, which will seek to improve the currently available models' shortcomings with respect to cyclic loading. Sandy soils are complex and therefore difficult to define fully with a constitutive model. Existing models prioritise modelling either static or cyclic loads at the expense of reducing accuracy with respect to the other load type. This project is novel in seeking to combine both cyclic and static loads into a single constitutive model.

This project is part of the Wind and Marine Energy Systems and Structures (WAMESS) Centre for Doctoral Training and falls within the EPSRC Wind Power, Ground Engineering and Energy research areas. This project will be run in partnership with Fugro, making use of their large commercial soil testing laboratory in Wallingford (near Oxford). The project will work in tandem with another doctoral project of a similar aim, but more focussed on mathematical and computational modelling aspects.